From a mathematical standpoint, the average could be pretty inaccurate (but might not be) as to the overall effect.
If we want to give a usable wind speed for the curve, then we need speed, direction, and some calculus to make it meaningful. You could have two different wind directions (and speeds, but ignore that), both of which will give the same measurement on the wind gauge, but the effect the winds have will be wildly different, and also depends on what lane you are in.